APROXIMACIONES EMPÍRICAS A LA INCIDENCIA DE LAS CONDICIONES

Cells respond to their environment by regulating their gene expression. An important aspect of molecular studies is the ability to detect differential gene expression patterns and define the profile of expression of particular genes. Clearly, if we are to fully understand the mechanisms that underlie tissue repair, we need to systematically define which genes are differentially expressed, and when, in a wound healing situation. H istorically many methods are available for detecting and quantifying the transcriptional levels of a gene. Some of these. Northern analysis, RNase protection and Reverse Transcriptase (RT-PCR) are methods for analysing the expression patterns and levels of known genes and so rely on having candidate genes for a particular study

Hybridisation, Serial Analysis of Gene Expression (SAGE) and Microarray analysis do not require knowledge of genes that are subsequently identified, and in theory allow a genome wide survey of wound-induced genes. All of these techniques are discussed below.

Northern Analysis

Northern analysis is the oldest and perhaps most widely used method for detecting and quantifying gene expression levels of particular known genes. RNA from a sample of interest, or from a series of different samples, is fractionated according to size by agarose gel electrophoresis and then blotted onto a nitrocellulose membrane for hybridisation. A radioactively-labelled, single stranded cDNA probe comprising a fragment of the gene being investigated is hybridised to the membrane where it binds to its corresponding sequence in the target RNA. The position of the resultant band on the blot gives information on the size of the investigated gene as well as revealing transcriptional information such as the presence of alternatively spliced transcripts. By carrying out densitometric studies on these blots the abundance of the transcript and thus total gene expression levels in the sample can be determined. Northern analysis was used to identify the transcriptional profiles of several known wound-induced genes such as CTGF and uPA that were differentially expressed during closure of wounds made to a renal epithelial cell monolayer (Pawar et al., 1995). However, Northern analysis is labour intensive and uses radioactivity and so this technique has largely been superseded by RNase protection and RT-PCR.

RNase Protection Assays

This assay involves the hybridisation of a radioactively-labelled probe complementary to the gene to be studied, to total RNA that has been extracted from a particular sample. This hybridised sample is then digested with an RNase cocktail that degrades single stranded RNAs. Those RNA species from the gene of interest that hybridise with the radioactive probe are double stranded and as such are ‘protected’ and so not degraded. Samples are separated by gel electrophoresis and like Northern analysis densitometry is used to obtain gene expression levels.

RNase protection assays have been used in several studies relating to tissue repair. The induction profiles of the three TGF(3 isoforms and the type I and type II receptors were revealed in an adult rat wound model by this assay (Frank et al., 1996). During murine cutaneous wound repair mRNAs for the pro inflammatory cytokines TN Fa and IL1(3 and a were found to be highly upregulated at an early stage (Hubner et al., 1996a) and a role for nitric oxide was indicated by the upregulation of Inducible Nitric Synthase

(Inos) and GTP Cyclohydrolase I (GTP-CHI) (Frank et al., 1998). RNase protection

assays have been instrumental in identifying several genes with roles in tissue repair but again this technique does involve the use of radioactivity and requires initial candidate genes, so cannot identify novel genes upregulated at the wound site.

RT-PCR and Real time PCR to determine gene expression

RT-PCR utilises the basic premise of standard genomic PCR but uses a cDNA derived from RNA as a template instead of DNA in order to determine expression levels of specific genes. Total or messenger RNA (mRNA) is extracted from the sample of interest and the reverse transcriptase enzyme used to synthesise cDNA. Target cDNA is then amplified with gene specific primers and the product run on an agarose gel to observe gene expression levels in the samples. Although standardly only semi quantitative, this technique can made more quantitative by incorporating fluorescent dyes, or the use of an internal competitor cDNA. Alternatively, quantification can be achieved with real time PCR that either incorporates a fluorescent dye into the reaction or uses the Taqman system (Gibson et al., 1996; Heid et al., 1996). In the case of fluorescent dye incorporation, a dye such as SYBR green is incorporated into the PCR reaction. The higher a gene is expressed in the sample of interest, the more cDNA template is available and so fewer PCR cycles need to be achieved before a sample is scored as statistically above background (Gibson et al., 1996). Inclusion of an internal control allows the relative intensity of the fluorescence and therefore the gene expression levels to be detected and quantified

The Taqman system (Applied Biosystems, UK) uses fluorescence to continually measure the PCR product accumulation. The RT-PCR reaction uses a dual labelled

of the gene to be amplified and has a reporter dye on the 5' end and a quencher dye on the 3' end. When the probe is intact, emission from the reporter is quenched by the quencher but during PCR amplification the probe is cleaved by the 5' nuclease activity of Taq polymerase separating the reporter from the quencher and allowing the fluorescence to be emitted. The system software examines the fluorescence intensity and plots increase in fluorescence over cycle number (time) producing a continuous measure of PCR amplification and, as such, a precise quantitative measure of gene expression.

A classic tissue repair experiment utilised an RT-PCR method to demonstrate the expression of TGFa and p, PDGF A-chain, and Insulin-like Growth Factor-1 by single cell macrophages isolated from murine wound fluid (Rappolee et al., 1988). As previously discussed, standard RT-PCR is not quantitative so more complex versions of this technique need to be employed to determine absolute gene expression levels.

Differential Display and Subtractive Hybridisation

Differential display uses PCR to detect genes that are differentially expressed between two samples. RNA is extracted from two samples to be compared and reverse transcribed with an oligo (dT) primer ending with C, G or A at the 3’ end splitting the resulting cDNA into three pools according to the final primer nucleotide. These cDNAs are amplified by PCR with a mix of random 5 ’ primers yielding several hundred PCR products corresponding to genes expressed in each sample. Separation by gel electrophoresis identifies bands present in one sample that are either absent from the other or that vary in intensity between the two as candidates for differentially expressed genes. These are then purified, subcloned and sequenced to determine the identity of the gene. Based on similar principles to differential display but much more sensitive and reliable, is subtractive suppression hybridisation where two samples are directly compared; the technique is made more sensitive to low abundance transcripts by equalizing their concentration with high abundance transcripts. Using hybridisation, sequences that are represented at the same level in both samples are subtracted out leaving only differentially expressed sequences that are then amplified by PCR. Again these have to be sequenced to determine identity.

These techniques have been particularly useful in identifying novel tissue repair genes. A large-scale screen was carried out by generating subtractive cDNA libraries to systematically identify genes that are differentially expressed 24 hours post wounding murine skin versus unwounded (Thorey et al., 2001). This isolated a number of novel and exciting genes not previously associated with wound healing, such as the SlOO family members, S100A8 (MRP8) and S100A9 (MRP14) (Odink et al., 1987). These were shown to be upregulated in wound kératinocytes and by invading inflammatory cells and were shown to associate with the keratin cytoskeleton of the cell, implying these genes might have a role in cytoskeletal rearrangements within the cell and a role in the initiation of chemotaxis of inflammatory cells to the wound site. Furthermore, the macrophage attractant, Chemokine 10 (also called MRP-1 or CCL6) was initially identified in this study and subsequently found to be strongly upregulated by kératinocytes and certain macrophages and is speculated to contribute to the maintenance of macrophage infiltration into the wound (Kaesler et al., 2002). Using a differential display approach to identify genes that were regulated by the growth factor KGF, a novel gene not expected to be active during wound healing was identified (Frank et al., 1997). Non-selenium glutathione peroxidase (NSGP) was shown to be highly upregulated by kératinocytes, particularly in psoriatic skin and subsequently shown to be expressed in embryonic and adult mouse wounds also (Grose, 1999) (Munz et al., 1997). NSGP is thought to protect cells at the wound site from oxygen toxicity, damage caused by reactive oxygen species that are known to be released at the wound site by inflammatory cells undergoing a protective oxidative burst (Halliwell and Gutteridge, 1990). The identification of NSGP as a wound gene demonstrates the ability of these techniques to identify novel genes and pathways that would not previously have been considered important in tissue repair. However, both differential display and subtractive hybridisation are prone to false positive errors and are not appropriate for saturation screening.

Serial Analysis of Gene Expression (SAGE)

defined region of mRNA that can uniquely identify a transcript. These short fragments are then joined together in concatamers that are sequenced; this sequence data then undergoes computational analysis to determine what genes these sequences correspond to. The frequency of each tag is a direct measure of abundance of corresponding mRNA and as such gene expression levels.

SAGE analysis has been used widely in the determination of differential gene expression and was used in a recent study to determine the genetic profile of cells responding to the Jun N terminal kinase (JNK) pathway which controls several morphogenetic movements (such as dorsal closure, a tissue movement analogous to wound closure - discussed earlier in the General Introduction), in the embryo of the fruit fly Drosophila melanogaster (Jasper et al., 2001). These authors identified many JNK responsive genes encoding both known and some novel cell adhesion molecules and cytoskeletal regulators in a SAGE comparison of embryos that were either genetically repressed or activated in the JNK signalling cascade. One important lesson from the Jasper et al analysis was the failure to identify several key JNK responsive genes, previously revealed by genetic studies. Almost certainly this was due to low expression levels of these genes, a general problem of gene expression profiling. Also, without further studies it is not possible using SAGE analysis to determine which gene inductions are directly due to JNK activation and which are downstream. One advantage of SAGE analysis is that virtually every mRNA expressed can be detected which is particularly practical for organisms with small genomes such as the fly genome but perhaps not so useful for higher organisms with a much larger genome size. SAGE analysis subsequently requires extensive bioinformatics, generally BLAST searches (http://www.ncbi.nlm.nih.gov/BLAST) to determine what genes the sequence tags making up the concatamers encode and many tags cannot be identified due to polymorphisms, sequencing errors or gaps in public sequence databases; for example, in the case of the Jasper et al study, 25% of tags were not identified. The quantity of sequencing involved in SAGE analysis and difficulties in reproducing protocols for concatamer formation have limited the use of this technique in studies of differential gene expression.

DNA Microarrays

A major limitation of most of the techniques discussed is that only a small number of genes can be analysed at one time, but the study of gene expression has been revolutionised in recent times by the development of DNA microarrays (Schena et al., 1995; Southern, 1975) that offer unprecedented large-scale throughput with the parallel probing of thousands of transcripts. Microarrays consist of a matrix of thousands of different DNAs (PCR products or oligonucleotides) representing known genes and cDNA expressed sequence tags (ESTs) spotted on to, or synthesised on, a solid support such as a glass slide. Technologies developed for constructing computer chips, allow thousands of genes to be spotted on a single slide. Hybridisation intensities for each DNA sequence on the array give a quantitative read-out of relative gene expression levels and can generate vast amounts of gene expression data simultaneously in a single hybridisation assay. Thus the technology provides a systematic and comprehensive way to study RNA and DNA variation in different situations. Over the past few years there has been a surge in interest in microarrays and the technology has been harnessed in a wider range of applications including gene expression profiling, comparative genomics and genotyping (Brown and Botstein, 1999; Hacia, 1999; Shoemaker et al., 1996; Wang et al., 1998). Even more widespread applications of the technology are anticipated. For example, expression analysis may facilitate the measurement of RNA levels for the complete set of transcripts of an organism. For DNA genotyping, an expansion to whole genome association studies determining genetic contribution to complex disorders is possible, and in mutation screening of disease genes, systematic and comprehensive testing of disease susceptibility in whole populations may become a reality (reviewed in (Lander, 1999).

There are two types of array, oligonucleotide arrays and cDNA arrays (Figure 1.1). For cDNA arrays, total or mRNAs are isolated from a pair of samples to be compared, labelled with different fluorescent dyes (typically Cy3 and Cy5) and hybridised simultaneously to the glass slide. The array is then washed and scanned for fluorescence intensities. Comparison of the intensities of the two dyes provides relative expression levels of thousands of genes in a single experiment. Oligonucleotide arrays,

the mouse there are three chips corresponding to 36,000 full-length mouse genes and ESTs, covering approximately the full mouse genome. Affymetrix Inc. manufacture chips for most of the model genomes including human, rat, fly, worm, yeast and even E.

coli. The enormous advantage of the Affymetrix GeneChip® over cDNA arrays is that

labelled mRNAs from different samples are hybridised to individual chips and the gene expression data from two or more samples can be compared using computer software to give a differential gene expression profiles. This allows one sample to be compared to more than one other, or compared to another at a later date.

S p o tte d DNA m ic ro a rra y High d e n s ity o lig o n u c le o tid e array DNA Public D a ta b a s e P C R am plification Purification R o b o tic printing G e n e X C o m b in ed im a g e in so ftw a re

/

^

' ÿ '

Ini mnaiii

G la ss ^ ■ ■ ■ ■ w afer O lig o m er s y n th e s is G e n e X RNA 1 + biotin RNA 2 + biotIn

\z:k

HybTMjize Im age 1 Im ag e 2 C o m b in ed d a ta in s o ftw are

Figure 1.4 Comparison of oligonucleotide and cDNA arrays

Diagram o f the m anufacture and assay of expression for oligonucleotide and cDNA arrays, (a) cD N A arrays are m ade by printing am plified genom ic cD N A s onto glass slides. Each spot corresponds to a contiguous gene fragm ent o f several hundred bases. O ligonucleotide arrays are made by light-directed com binatorial chem ical synthesis o f thousands o f highly ordered oligom er probes onto the glass chip. Each gene is represented by 15-20 different oligom er pairs (PM , perfect match and MM , m ismatch), (b) For cD NA arrays, m RNAs from a sam ple of interest ( I ) and a reference (2) are labelled with different fluorophores, typically Cy3 and Cy5, hybridised to the slide and scanned to detect both fluorophores. X represents a gene with increased expression level in sample 1, Y, increased levels in sam ple 2 and Z represents no difference betw een sam ples I and 2. W ith oligonucleotide arrays such as the GeneChip®, RNA is labelled to produce biotinylated cRNA and hybridised to the array that is scanned with a fluorophore conjugated to avidin and detected by laser scanning. X represents paired oligonucleotides fo r a gene present at increased levels in sam ple I, Y, increased levels in sam ple 2 and Z represents sim ilar expression levels in sam ples I and 2. Diagram adapted from (Harrington et a i, 2000)

Uses of microarray-based expression profiling

The power of this technology has already been applied to a wide range of physiological situations, for example, in cancer classification (Golub et al., 1999) (Alizadeh et al., 2000), identification of human disease genes (Lawn et al., 1999) and in the analysis of global gene expression in single-celled organisms (Chu et al., 1998). It has also been applied with particular success to analyse the dynamic gene expression profile during the development of Drosophila melanogaster (White et al., 1999). This particular study systematically identified the genetic profile of Drosophila metamorphosis, a complex and dramatic process involving cellular proliferation, tissue remodelling, cell migration and programmed cell death in a developmental process coordinating the action of hundreds of genes. Using a cDNA array containing 4500 Drosophila genes and by interpreting the data by grouping genes according to the similarity of their expression profiles (a method that will be discussed further in Chapter 4), they were able to assign many differentially expressed genes to developmental pathways, as well as pathways not previously associated with, metamorphosis. This approach demonstrates the ability of microarray technology to study complex biological processes in multicellular organisms. Indeed, many of the gene cascades identified in this study may be re­ activated at the wound site. As discussed in the General Introduction, morphogenetic tissue movements in simple organisms such as dorsal closure in Drosophila appear to be analogous to the tissue movements of wound closure with both movements dependant on JNK signalling and using similar actin machineries (Ramet et al., 2002) (Wood et al., 2002).

Conversely, Iyer et al (1999) used DNA microarrays to identify the gene expression profile of cell populations in tissue culture. They used a cDNA array containing 8600 human genes and identified 517 that were differentially expressed when fibroblasts were exposed to serum, a physiological response model in the study of growth control and cell cycle progression. This provided the first detailed picture of gene expression induced by a growth factor stimulus, much as occurs in vivo at the site of tissue damage. As such, this study was promoted as revealing the transcriptional response to wounding and indeed, many of the gene profiles in their study are recapitulated in my own study.

situation is the investigation of the genetic profile of murine pulmonary fibrosis, a condition triggered by a chronic inflammatory response (Kaminski et al., 2000). Inflammation was induced in the lung by administration of bleomycin and the genetic

In document Costeo de la cobertura de agua potable y alcantarillado a nivel cantonal (página 40-46)